A Neighborhood Gradient Based Infrared Dim Target Detection Method in Complex Background

Abstract

The invention discloses a method for detecting dim and small infrared targets in complex backgrounds based on neighborhood gradients, comprising: 1) generating initial search main window and sub-window templates; 2) obtaining the foreground judging window pixel matrix according to the minimum value of the window; 3) constructing the foreground Judging the mapping table; 4) Calculating the neighborhood peak gradient feature and neighborhood step gradient feature; 5) Judging the SVM classification weight of the window feature; 6) Judging the neighborhood contrast of the window; 7) Adaptively adjusting the window step size , complete single-frame image traversal and output target information; 8) Continuous target detection is performed on sequence images, and motion characteristic judgment and static characteristic judgment are performed respectively, and the target detection information data of the latest frame is calculated. The present invention can be used in scenarios such as remote photoelectric reconnaissance, unmanned aerial vehicle detection, coastal defense, vehicle-mounted object reconnaissance and other scenarios, and solves the problem of low detection accuracy, high false alarm rate, and low detection robustness in the weak and small target detection technology under the existing complex background The problem.

Description

technical field [0001] The invention relates to the technical field of digital image processing, in particular to a method for detecting infrared weak and small targets based on neighborhood gradients. Background technique [0002] As one of the research hotspots of photoelectric image processing technology, infrared faint target detection is widely used in the field of early warning and reconnaissance. Infrared weak and small targets have the characteristics of small imaging size, weak shape features, unobvious texture features, and are susceptible to noise interference, especially in complex backgrounds such as clouds, ground objects, and sea surfaces, which are likely to cause false and missed alarms. Therefore, in order to improve the performance of the photoelectric reconnaissance system, it is of great practical significance to study how to detect weak targets in complex backgrounds. [0003] Yang Hui et al. proposed a background difference detection algorithm that co...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the deficiencies of the existing technology, the present invention proposes a detection method for infrared weak and small targets in complex backgrounds based on neighborhood gradients in order to solve the problems of detection effectiveness and robustness of small infrared targets in complex backgrounds. Low computational complexity and easy engineering implementation

Images